Effects of High Temperature and Different Salt Solutions on Basalt Fiber-Reinforced Composites’ Bonded Joint Durability Impact

: Fiber-reinforced composites are widely used in industrial development due to their excellent performance, and the study of basalt ﬁber-reinforced resin (BFRP) as a new type of economical and environmentally friendly material is highly valued, since harsh environments can affect the durability of bonded joints. In this paper, the Araldite ® 2015 adhesive for BFRP–BFRP single lap joints (SLJs) was selected as the subject of study and the joints were analyzed in aging experiments in three environments: deionized water (DW), 3.5% NaCl solution, and 5% NaCl solution at 80 ◦ C for 0 days (no aging), 10 days, 20 days, and 30 days. Using Fick’s second law to describe water absorption in joints and materials, the comparison shows that the water absorption in the joints occurs primarily in the adhesive. Differential scanning calorimetry (DSC) was used to characterize the decrease in the glass transition temperature (Tg) of the adhesive at each failure point, and the thermogravimetric analysis (TGA) tests showed that moisture and heat led to the degradation of the polymer material in the joint. The failure strength of the joints in quasi-static tensile tests was positively correlated with the moisture content of the solution, and the changes in the absorption peaks of the functional groups of the adhesive after aging were observed. The comprehensive macro-micro failed section analysis showed that the water molecules damage the chemical properties of the adhesive, meaning that the adhesive and BFRP binding ability is decreased. The proportion of failure section tear decreased with the extension of the aging time, and a high temperature induced water evaporation and an adhesive post-curing reaction. The change in the failure mode is a result of the combined effect of the post-curing effect and hydrolysis reaction, which is validated by the results of the Fourier infrared spectroscopy (FTIR). This study contributes to an in-depth understanding of the effect of moisture and heat on the residual properties of bonded joints.


Introduction
The concept of green development places higher demands on the design of car body structures.Relevant studies have shown that the lightweight design of car bodies can reduce the energy consumption of automobiles.Therefore, the demand for lightweight components in industries such as aerospace, shipbuilding, automobiles, and railways is constantly increasing [1].The use of lightweight materials has become a trend in the future development of car body structure design.Fiber composite materials have entered researchers' field of vision due to their superior mechanical properties, light weight, and other advantages.These include carbon fiber-reinforced composites (CFRPs), glass fiberreinforced composites (GFRPs) and basalt fiber-reinforced composites (BFRPs) [2].A new research hotspot in recent years, BFRPs not only have the characteristics of fiber-reinforced composites, but also differ from the commonly used CFRPs and GFRPs in that the basalt fibers are made from molten basalt rock, with a wide range of material sources, a less energy-intensive manufacturing process, no additives, excellent recyclability, environmental friendliness, and lower cost.Vedernikov A et al. [3] investigated the flexural and interlaminar shear properties of 0 • and 90 • fiber orientations determined by increasing the pultrusion speed of the pultrusion process.The differences in the mechanical properties of structural composites of glass fiber/vinyl ester resins can be explained by the presence of air bubbles, longitudinal voids, and matrix cracks, and by the fact that their density and dimensions increase with an increasing pultrusion speed.Wensu Chen et al. [4] conducted quasi-static and dynamic tests on BFRPs to analyze and discuss the strain rate sensitivity of BFRP materials.When the strain rate exceeds 120 s −1 , the BFRPs' properties, including tensile strength, failure strain, and elastic modulus, rapidly increase with the strain rate.V. Lopresto et al., [5] by comparing the test results of two types of laminates, glass and basalt fiber-reinforced plastics, found that the basalt material has high properties in terms of Young's modulus, compressive and flexural strength, impact, and energy.These good properties indicate that basalt fibers have the potential to be used in areas where glass composites are currently widely used.Jianzhe Shi et al. [6] found that BFRPs have an increase in strength and modulus of about 20% compared to GFRPs at a similar cost and higher chemical stability; a wider range of operating temperatures and a much lower cost than CFRPs; and the residual strength of BFRPs after 1000 h of sustained loading still reached about 95% of its initial tensile strength.
In automotive engineering, the joining techniques between BFRPs and other materials in the structure, as well as between materials of the same type, limit its further development.Traditional bolting, riveting, and welding not only damage the original structure and surface of the composite, but also inevitably cause problems such as fiber breakage and stress concentration.Bonding technology has now become a widely accepted composite joining process, and, as a new type of joint, it has the advantages of uniform stress distribution, good bonding performance, light weight, and so on.Junjia Cui et al. [7] studied the effect of the rivet layout on the mechanics of bonded CFRP-Al joints.The results showed that applying four rivets in a square array to bonded joints improved the shear strength of the joints.Observing the cross-sectional area, it was found that the addition of rivets did not alter the failure mode of the joint.G. Marannano et al. [8] used rivets of different materials to provide different effects on the life of the bonded CFRP-Al joints.The mixed joint life of steel rivets was higher.At the same time, the author also found that riveting leads to the local layering and stress concentration of CFRP materials.For anisotropic composite materials, this damage reduces the load-bearing capacity of the joint, and adhesive joints can maximize the protection of the composite material, resulting in a more uniform load distribution.Anastasios P. Vasilopoulos et al. [9] investigated the tensile behavior of GFRP bonded double ring joints between 35 • C and 60 • C. The DMA test results indicate that the GFRP material itself is less affected by temperature, and the decrease in the mechanical properties of the joint is mainly due to the aging behavior of the adhesive under conditions higher than the glass transition temperature.Isaiah Kaiser et al. [10] explored a biomimetic pattern adhesive based on biological inspiration for CFRP-Ti single lap adhesive joints.The results showed that the fracture path of the adhesive joint continuously reorients along the geometric characteristics of the gecko foot, resulting in higher energy absorption when the joint fractures.Although adhesive joints have potential advantages such as a high strength to weight ratio, design flexibility, and ease of manufacturing, automobiles and multiple units typically operate in high temperatures or corrosive environments, and the mechanical properties after the long-term exposure to harsh environments are still a problem [11][12][13].
Coatings 2023, 13,1936 3 of 21 Many scholars have studied the effect of temperature on the stability and glass transition temperature of different types of adhesives and concluded that the adhesive is exposed to high temperatures in which van der Waals bonds break [14] and side group reactions occur within the adhesive [15], leading to changes in the stability of the adhesive.The effect of each mechanical parameter of composites exposed to different temperatures is different [16]; therefore, temperature and humidity are important factors affecting the loadcarrying capacity of bonded joints.Adhesive joints absorb water, resulting in irreversible losses due to plasticization, hydrolysis, and expansion [17].In corrosive environments such as seawater, metal corrosion leads to more severe damage to adhesive joints [18].M. Bordes et al. [19] investigated the long-term strength of bonded steel/epoxy joints in seawater, determined the relationship between the adhesive moisture content and adhesive performance loss, and analyzed the stress state of bonded joints based on the relationship between the two.G. C. Papanicolaou et al. [20] investigated boron/epoxy single lap joints in salt spray environments.The entry of water molecules in the salt spray environment resulted in a plasticizing effect at the edges of the joints, resulting in a decrease in the joint's load-bearing capacity.The plasticizing area continued to expand over time, ultimately leading to joint failure.Secondly, the lap length is an important factor affecting the loadbearing capacity of the joint.When the lap length reaches 13 cm, the cost-effectiveness of the bonded joint reaches its best.Benjamin J. Anderson [21] suggests that the degradation of adhesive networks leads to a decrease in adhesive strength.In addition, the relationship between the environmental temperature and Tg value also affects the adhesive strength of the adhesive [22].Jingxin Na et al. [23] investigated the effect of temperature on the mechanical properties of BFRP-Al joints in the automotive industry.Under high temperature (ambient temperature higher than Tg) conditions, the increased ductility of the adhesive resulted in a decrease in the Young's modulus and load-bearing capacity of the joint; when the environmental temperature is lower than the glass transition temperature of the adhesive, the adhesive is in a glass state, so the bonding joint is less affected by temperature.Budhe et al. [24] found that the hydrothermal coupling effect accelerates the aging process of bonded joints compared to a single factor.Wenlong Mu et al. [25] exposed the CFRP-Al joint to an environment of 80 • C-95% RH.After aging, due to the humid and hot environment, the chain cracking, crosslinking density, and glass transition temperature of the adhesive decreased by 13.14%, and Araldite ® 2015 is more sensitive to damp heat aging than CFRP.Gerald Doyle et al. [26] investigated the bonding of aluminum joints and exposed the joints to a 65 • C deionized water environment.The results showed that during the aging process, water molecules entered the interior of the adhesive, leading to the plasticization of the adhesive and a decrease in the Tg value, as well as the surface corrosion of the aluminum substrate, ultimately leading to the failure of the bonding joint.N.S.Hirulkar et al. [27] found that after 60 cycles of exposure to moist heat aging combined with thermal shock in air, the breaking loads of ductile adhesives were reduced by about 12.50% and 12.5%, respectively, in comparison to the breaking loads of brittle adhesives, which were about 38.6% and 24.8%, respectively.Regardless of which adhesive is used, higher hygrothermal aging temperatures result in a sharp decrease in the breaking load and ultimate strain.The breaking load and ultimate strain were minimally affected by thermal shock in air, whereas the breaking load decreased significantly with increasing damp heat shock.
The service environment temperature of automobiles is about −40-80 • C. The mechanical properties of adhesives and BFRP composites will change in a continuous hot and humid environment, resulting in the degradation of bonding structure properties.Little work has been performed on investigating the mechanical properties of BFRP bonded joints, so it is essential to investigate the effects of high temperatures and different salt solutions on the durability of basalt fiber-reinforced composite bonded joints to provide automotive manufacturers with advice on material selection, design, and the conditions of use to ensure that the bonded joints perform consistently in a variety of environments.In this paper, Araldite ® 2015 adhesive was selected to bond BFRP-BFRP single lap joints, and the thermal and mechanical properties of each joint were tested after aging in three environments: deionized water (DW), 3.5% NaCl solution, and 5% NaCl solution at 80 • C for 0 days (no aging), 10 days, 20 days, and 30 days, respectively.This study involved the use of Fick's second law to describe the water absorption of the joints and materials, the analysis of the aging characteristics of adhesives and BFRP composites by DSC and FTIR tests, TGA-DTG experiments to analyze the thermal stability of the adhesives in different aging environments, the comparison of the change rule of the failure load of the joint performance in different aging environments, the observation of the failure fracture morphology of each joint, and the analysis of the failure mode of the joints.This was to provide support for the data for the application of BFRP in the automobile manufacturing industry.

Materials
The BFRP (Zhongdao Technology Company, Jilin, China) selected for the study consists of an epoxy resin matrix filled with basalt fibers processed through plain weave and prepreg, basalt fibers (Zhongdao Technology Company, Changchun, Jilin Province, China) adopt a [0/90/0/90/0/90/0/90] layup, with a monofilament diameter of 13 µm, and the composite molding resin is 5113-81A (epoxy resin).The composite molding resin consists of 5113-81A (epoxy resin) and 5113-94B (curing agent), two components, and its ratio is 100:25, the performance parameters are shown in Table 1, and the fiber volume ratio is 65%-70%.Connecting the substrate is the selected adhesive Araldite ® 2015 (Huntsman Advanced Materials, Guangzhou, Guodong Province, China), a two-component epoxy system with amine curing agent and DGEBA-based epoxy resin.Shear and peel strength is high, with thixotropic and low shrinkage characteristics, a glass transition temperature (Tg) average value of 75 ± 4 • C. Since curing occurs at room temperature, there is a longer working time; a 40 min application time.It is a tough adhesive with high shear strength and peel strength.A total of 4 h is required for initial curing, and 6 h to move.It has a high hardness after curing, and low shrinkage.For material performance parameters, see Table 2.

Preparation of Test Pieces
BFRP-BFRP single lap joints were fabricated, and the bonding area was 25 cm × 25 cm.The thickness of the adhesive layer is controlled by glass beads with a diameter of 0.1 mm.The single lap joint is shown in Figure 1.The production process of the joint is divided into the following steps.
the surface, and maximize the effect of the adhesive.
(5) Apply glue.After wiping with acetone for 10-15 min, use glue gun to glue.( 6) Carry out bonding.The whole bonding process is carried out at room temperatu (25 ± 2 °C).When bonding, lap two BFRP substrates on the fixture, add gaskets control the thickness of the adhesive layer at 0.1 mm, and finally press and tight the joints.

Experimental Method
After the specimen preparation, the specimens were placed in a high and low te perature humidity and heat alternating test chamber (WS-1000, Weiss Equipment Lab atories, Inc.) and aged at 80 °C (deionized water, 3.5% NaCl and 5% NaCl) for 0 days ( aging), 10 days, 20 days, and 30 days, respectively.The specimens were taken out fro the experiment box after they reached the required time, and then the quasi-static tens test using the electronic universal tensile test machine (China Jinan Xinguang Testing M chine Manufacturing Co., Ltd., Jinan, China) was performed.After they were dried room temperature, a loading rate of 2 mm/min and the load-displacement curve and t maximum failure load of the specimen were obtained.In addition, in order to preve bending from occurring during the stretching process, shims with a thickness of 2 m were placed at both ends of the specimen.In order to ensure that the experimental resu were real and reliable, three repetitions of the experiments were carried out and the o tained experimental data were averaged, as shown in Figure 2. Through the control co puter connected to the universal tensile machine, the average failure strength of ea (1) Selection of BFRP substrate.In order to ensure the area of the bonding area, the water jet is used to accurately cut the sheet, and the cut sheets are reasonably matched.(2) Numbering the test pieces.In order to avoid obtaining the wrong substrate during the bonding process, the test pieces are numbered.(3) Mark the bonding area.Mark a horizontal line at 25 cm on the BFRP substrate to ensure that the bonding area is 25 cm × 25 cm.(4) Wipe the BFRP substrate: Since the BFRP substrate will stick to grease and dust during the production process, this can seriously affect the bonding performance, meaning that the research cannot draw the most realistic conclusion.Therefore, use acetone to wipe the surface of BFRP before bonding to remove the grease and dust on the surface, and maximize the effect of the adhesive.(5) Apply glue.After wiping with acetone for 10-15 min, use glue gun to glue.( 6) Carry out bonding.The whole bonding process is carried out at room temperature (25 ± 2 • C).When bonding, lap two BFRP substrates on the fixture, add gaskets to control the thickness of the adhesive layer at 0.1 mm, and finally press and tighten the joints.

Experimental Method
After the specimen preparation, the specimens were placed in a high and low temperature humidity and heat alternating test chamber (WS-1000, Weiss Equipment Laboratories, Inc.) and aged at 80 • C (deionized water, 3.5% NaCl and 5% NaCl) for 0 days (no aging), 10 days, 20 days, and 30 days, respectively.The specimens were taken out from the experiment box after they reached the required time, and then the quasi-static tensile test using the electronic universal tensile test machine (China Jinan Xinguang Testing Machine Manufacturing Co., Ltd., Jinan, China) was performed.After they were dried in room temperature, a loading rate of 2 mm/min and the load-displacement curve and the maximum failure load of the specimen were obtained.In addition, in order to prevent bending from occurring during the stretching process, shims with a thickness of 2 mm were placed at both ends of the specimen.In order to ensure that the experimental results were real and reliable, three repetitions of the experiments were carried out and the obtained experimental data were averaged, as shown in Figure 2. Through the control computer connected to the universal tensile machine, the average failure strength of each specimen could be read, and the obtained data could be processed and plotted to obtain the changes in the average failure strength of the joints in each environment.Also, the load-displacement curves and the analysis of the image changes can be summarized to show the relationship between the aging environment and the aging time on the mechanical properties of the joints.Macro-analysis of the failure section of the joint combined with SEM images can be derived from the failure mode of each aging environment at different aging nodes.

PEER REVIEW 6 of 22
specimen could be read, and the obtained data could be processed and plotted to obtain the changes in the average failure strength of the joints in each environment.Also, the load-displacement curves and the analysis of the image changes can be summarized to show the relationship between the aging environment and the aging time on the mechanical properties of the joints.Macro-analysis of the failure section of the joint combined with SEM images can be derived from the failure mode of each aging environment at different aging nodes.

Water Absorption Test
In order to more clearly and intuitively analyze the role of moisture in the aging of joints, dumbbell-type standards of Araldite ® 2015 were prepared for water absorption testing with reference to ISO527 [28], with geometric parameters, as shown in Figure 3. Three standards were taken from each group, and in accordance with ASTM D570-98 (2018) [29], the water absorption rate was measured by the gravity method at regular intervals of 24 h between each group of standards.The weight was measured, the water absorption was calculated by comparing the initial mass, and then the obtained data were averaged.The weight measurement used an analytical balance with an accuracy of 0.1 mg, and before weight measurement, absorbent paper was used to wipe away and remove the surface moisture.The weight measurement process is not more than 30 min, in order to avoid the influence of the external environment on the experimental results.The formula is shown in Equation ( 1).
% 100 where t M denotes the water absorption at time t, denotes the mass at time t, and 0 W denotes the original mass.

Water Absorption Test
In order to more clearly and intuitively analyze the role of moisture in the aging of joints, dumbbell-type standards of Araldite ® 2015 were prepared for water absorption testing with reference to ISO527 [28], with geometric parameters, as shown in Figure 3. Three standards were taken from each group, and in accordance with ASTM D570-98 (2018) [29], the water absorption rate was measured by the gravity method at regular intervals of 24 h between each group of standards.The weight was measured, the water absorption was calculated by comparing the initial mass, and then the obtained data were averaged.The weight measurement used an analytical balance with an accuracy of 0.1 mg, and before weight measurement, absorbent paper was used to wipe away and remove the surface moisture.The weight measurement process is not more than 30 min, in order to avoid the influence of the external environment on the experimental results.The formula is shown in Equation (1).
where M t denotes the water absorption at time t, denotes the mass at time t, and W 0 denotes the original mass.

DSC Test
Glass transition temperature (Tg) is an inherent property of adhesives, BFRP, and other polymer materials, and is the macroscopic embodiment of the changes in the form of molecular motion, which directly affects their mechanical properties in different temperature environments.When the temperature is below this temperature, the material exhibits rigidity in the glassy state, and when the temperature is above this temperature the material exhibits elasticity in the highly elastic state.Tg depends on many factors such as molecular weight, chemical cross-linking, plasticizers, polymerization chain length, and functional groups [30], and its alteration is one of the main factors that lead to the aging degradation of the joint properties, with a non-negligible effect of temperature and saline water [31].In order to be able to more accurately characterize the role of Tg in the aging of the joints, the experiments were carried out with a differential scanning calorimeter (Mettler Toledo, DSC3+, Zurich,Switzerland) for DSC analysis of the adhesive Tg at different aging and aging nodes.
DSC utilizes the principle of dynamic zero equilibrium, which allows the temperature difference between the sample and the reference to converge to zero to obtain the desired energy difference.The tests were conducted in a nitrogen atmosphere with a ramp-up/down rate of 5 °C/min and a temperature range of −70 to 200 °C.The adhesive sample mass was approximately 5 mg from a failed section of the bonded joint.Two warm-ups were required for each test, the first being used to remove the thermal history of the sample, and the Tg was determined from the second warm-up process.

FTIR Test
Fourier transform infrared spectroscopy (FTIR) uses the absorption properties of substances to different wavelengths of infrared radiation, a method for molecular structure and chemical composition analysis.FTIR testing of the adhesive samples after aging treatment was performed to obtain the FTIR spectra of the adhesive before and after aging.Using a VERTEX 80V Fourier Transform Infrared Spectrometer, the IRATR spectrum of the failure section of the BFRP-BFRP joint was obtained, and the penetration of infrared radiation into the sample was avoided using the totally attenuated multiple reflection method (IRATR).The spectrometer platform was purged with dry nitrogen prior to

DSC Test
Glass transition temperature (Tg) is an inherent property of adhesives, BFRP, and other polymer materials, and is the macroscopic embodiment of the changes in the form of molecular motion, which directly affects their mechanical properties in different temperature environments.When the temperature is below this temperature, the material exhibits rigidity in the glassy state, and when the temperature is above this temperature the material exhibits elasticity in the highly elastic state.Tg depends on many factors such as molecular weight, chemical cross-linking, plasticizers, polymerization chain length, and functional groups [30], and its alteration is one of the main factors that lead to the aging degradation of the joint properties, with a non-negligible effect of temperature and saline water [31].In order to be able to more accurately characterize the role of Tg in the aging of the joints, the experiments were carried out with a differential scanning calorimeter (Mettler Toledo, DSC3+, Zurich, Switzerland) for DSC analysis of the adhesive Tg at different aging and aging nodes.
DSC utilizes the principle of dynamic zero equilibrium, which allows the temperature difference between the sample and the reference to converge to zero to obtain the desired energy difference.The tests were conducted in a nitrogen atmosphere with a ramp-up/down rate of 5 • C/min and a temperature range of −70 to 200 • C. The adhesive sample mass was approximately 5 mg from a failed section of the bonded joint.Two warm-ups were required for each test, the first being used to remove the thermal history of the sample, and the Tg was determined from the second warm-up process.

FTIR Test
Fourier transform infrared spectroscopy (FTIR) uses the absorption properties of substances to different wavelengths of infrared radiation, a method for molecular structure and chemical composition analysis.FTIR testing of the adhesive samples after aging treatment was performed to obtain the FTIR spectra of the adhesive before and after aging.Using a VERTEX 80V Fourier Transform Infrared Spectrometer, the IRATR spectrum of the failure section of the BFRP-BFRP joint was obtained, and the penetration of infrared radiation into the sample was avoided using the totally attenuated multiple reflection method (IRATR).The spectrometer platform was purged with dry nitrogen prior to analysis.The average resolution was 4 cm −1 and the spectral range was 4000-400 cm −1 for the 200 scans.

TGA-DTG Analysis
The thermogravimetric method (TGA) is a thermal analysis technique for the relationship between the mass of the sample to be tested and the temperature or time change under a program-controlled temperature and a certain environment.Component analysis and qualitative analysis of adhesives can be carried out by thermogravimetric analysis.In the experiment, the thermal stability behavior of the adhesive was investigated using a TGA PerkinElmer Pyris 1 TGA thermogravimetric analyzer (USA).The experimental environment was as follows: nitrogen environment, the temperature range was 25 • C-800 • C, and the heating rate was 10 • C/min.The weight loss rate St of the sample by temperature was analyzed, and calculated as in Equation ( 2):

Hygroscopicity Analysis
Adhesives absorb moisture in hot and humid environments, which deteriorates the adhesive thermo-mechanical properties (Tg, modulus of elasticity, yield strength) and adhesion.In the aging environments of 80 • C-DW, 80 • C-3.5% NaCl solution, and 80 • C-5% NaCl solution, moisture gradually diffuses into the adhesive, which can be viewed as a rectangular model of the adhesive.
When the material is isotropic, Equation (3) can be simplified, viz.: In Equation (3), C denotes the diffusion concentration of the substance, D denotes the diffusion coefficient of the substance, t denotes the diffusion time, and x, y, and z denote the diffusion along the x, y, and z directions, respectively.
For thin plate-like materials, whose thickness and width are much smaller than their length, diffusion can only be considered in the x-direction, simplifying the equation as: Equation ( 6) C 0 Indicates the moisture content of the material at the initial moment, C ∞ represents the moisture content in the material when saturated.When C 0 and C ∞ are equal, the hygroscopicity of the specimen reaches dynamic equilibrium.h represents the specimen thickness of 1/2.
By integrating Equation ( 6), the relationship between the suction volume and saturated suction volume at any moment can be obtained as Coatings 2023, 13,1936 9 of 21 M 0 represents the percentage of moisture content of the specimen at the initial moment, M ∞ represents the percentage of moisture content of the specimen at saturation, D is the diffusion coefficient, and M t can be calculated by Equation (8).
Theoretically, the Fickian diffusion process is usually characterized by a square root curve of moisture absorption with time, and the equation can be approximated by the following expression: In the course of the experiment, M ∞ can be measured experimentally.For the calculation of D, it is assumed that the specimen is infinite and that there is no edge diffusion, which in fact occurs in all six rectangular sections.A correction factor to account for the edge effects can be derived so that the true 1D diffusion coefficient D is expressed as The diffusion coefficient D is temperature-dependent and follows the Arrhenius relation for an ideal system, viz.: In Equation ( 11), E α represents the activation energy during diffusion, R represents the ideal gas constant, and T represents the absolute temperature.
Adhesives absorb moisture in a humid and hot environment, leading to the deterioration of the thermal mechanical properties (Tg, elastic modulus, yield strength) and adhesion of the adhesive.
The diffusion coefficient and saturation moisture absorption of the Araldite ® 2015 adhesive and BFRP substrate in different concentrations of NaCl solution environment are shown in Table 3.The basalt fibers of the BFRP substrate did not participate in the water absorption process, resulting in a much lower diffusion coefficient and saturation moisture absorption rate of the BFRP substrate than the adhesive.The diffusion coefficient of the adhesive in the deionized water environment is 3.53 × 10 −3 .When the concentration of NaCl solution increases, the osmotic pressure in the solution increases [32], accelerating the movement of water molecules towards the interior of the adhesive, thus increasing the diffusion coefficient.Specifically, the saturated water absorption of the adhesive decreases with increasing ion concentration.By comparing the water absorption of the adhesive and the BFRP in the three environments, it can be found that increasing the temperature promotes the diffusion of water into the interior of the material, and increasing the salt concentration under normal seawater salt concentration inhibits the water absorption of the material, which is related to the penetration effect of the salt ions.The results of the water absorption test reflect the extent to which the materials of each part of the structure may be affected by moisture in each case.Comparing the water absorption of the two materials of the joints, it can be found that the water absorption of the adhesive is significantly higher than that of the BFRP in the three environments, and although both the adhesive and the BFRP have epoxy resin compositions, the fibers in the BFRP are almost non-absorbent, which results in the BFRP having a much lower water absorption, and this leads to the conclusion that the damage caused by moisture in the joints mainly occurs on the level of the adhesive, and that hydrolytic damage to the BFRP can only be more clearly observed under high temperature environments.The adverse effects of moisture after entering the material, physical swelling and plasticization, are reversible, while the irreversible effects at the molecular level are mainly twofold: on the one hand, chemical reactions lead to the entanglement of the internal pores of the polymer chains, ultimately leading to the formation of microcracks in the material body and the weakening of the bonding interface.On the other hand, water molecules interact with polar groups, in which absorbed water continuously collides with the crosslinked chains in the adhesive, leading to chain breaks and segment leaching.The much greater water absorption of the adhesives than the BFRP may be related to the functional groups in the molecular chains [33], as the reaction with the functional groups after the water molecules have filled the free volume is a key factor in driving the material to absorb more water.

Adhesive DSC Results
The DSC test can determine the extent of chemical reactions in adhesives during the aging process [34], and several researchers have previously explored the Tg values of adhesives during the aging process.The results show that the greater the molecular chain flexibility, the lower the Tg; the greater the molecular chain rigidity, the higher the Tg value [35].The glass transition temperature is an important parameter in determining the properties of adhesives and BFRP composites and is specified as the Tg reference value in a wide range of applications.The Tg values of the adhesive Araldite ® 2015 in different aging environments are shown in Figure 4, and its changes in different aging environments are shown in Figure 5.
From Figure 4, it can be seen that with the aging time growth in each environment of the adhesive, the Tg decreases slightly.The reason for this may be the long period of exposure to high levels of humidity that causes an adhesive and resin matrix plasticizing effect and chemical modification, causing the Tg to decrease.Humid and hot environments may result in local molecular chain breakage and the crosslinking density being reduced, which leads to the reduction in the Tg.When not aged, the Tg value of the adhesive is 63.77 • C.After aging, the Tg values of the adhesive showed varying degrees of decrease in all three environments.In a deionized water environment, after 30 days of humid heat aging, the Tg values were all 34.17 • C, showing a decrease of 46.42%.The Tg values in the salt solution environment decreased by 18.19% and 15.73%, respectively.This is due to the absorption of moisture by the adhesive in hot and humid environments, which leads to molecular chain breakage and a decrease in cross-linking density, which is manifested as a decrease in Tg value [36].In particular, Figure 5 shows that the Tg value of the adhesive decreases the fastest in the deionized water environment and the slowest in the 5% NaCl solution environment.Combined with the analysis of the hygroscopicity of the adhesive, it can be concluded that moisture is an important factor affecting the performance of the adhesive, and that a greater amount of water absorbed by the adhesive leads to a greater change in the chemical properties of the adhesive.This is because the water absorption analysis shows that water is absorbed by polymers in two different ways: as free water, which occupies the free space of the polymer and leads to the plasticization of the material, which can lead to a decrease in the Tg; and as bound water, which forms hydrogen bonds with hydrophilic groups such as hydroxyl and amine groups in the polymer network, which can lead to an increase in the Tg.The hydrophilicity of the adhesive was stronger than that of the BFRP in the water absorption test, suggesting that there are more hydrophilic groups in the adhesive, i.e., the binding water absorption will be more pronounced, attenuating the decreasing trend of the Tg.performance of the adhesive, and that a greater amount of water absorbed by the adhesive leads to a greater change in the chemical properties of the adhesive.This is because the water absorption analysis shows that water is absorbed by polymers in two different ways: as free water, which occupies the free space of the polymer and leads to the plasticization of the material, which can lead to a decrease in the Tg; and as bound water, which forms hydrogen bonds with hydrophilic groups such as hydroxyl and amine groups in the polymer network, which can lead to an increase in the Tg.The hydrophilicity of the adhesive was stronger than that of the BFRP in the water absorption test, suggesting that there are more hydrophilic groups in the adhesive, i.e., the binding water absorption will be more pronounced, attenuating the decreasing trend of the Tg.

FTIR Results
In order to investigate the changes that occur in the molecular groups of the material during the aging process of the adhesive at high temperatures, the adhesive was tested by Coatings 2023, 13,1936

FTIR Results
In order to investigate the changes that occur in the molecular groups of the material during the aging process of the adhesive at high temperatures, the adhesive was tested by Fourier transform infrared spectroscopy (FTIR), and the results of the test are shown in Figure 6.In the curves of Figure 6, curve a0 represents the unaged curve, a1~a3 represent the deionized water aging environment (10 days, 20 days, and 30 days), a4~a6 represent the 3.5% NaCl aging environment (10 days, 20 days, and 30 days), and a7~a9 represent the 5% NaCl aging environment (10 days, 20 days, and 30 days).The absorption characteristics of the samples before and after aging can be observed in the range of 400~4500 cm −1 , and different absorption peaks represent different chemical groups, which can be used for qualitative analysis.The functional groups corresponding to the characteristic peaks of the IR spectra of the adhesive are shown in Table 4.
The process in which molecules enter the epoxy resin and react with the corresponding hydrophilic groups is known as hydrolysis, which not only destroys the molecular chain of the polymer in the epoxy resin adhesive, but also destroys the cross-linking of the ester groups, leading to the generation of new carbonyl and alkyl groups.Prolonged moisture and heat attack can lead to changes in the chemical composition of the adhesive, causing irreversible damage to the material.In addition, the lower intensity of the vibrational peaks for carbonyl and alkyl groups with increasing aging time indicates that the postcuring reaction hinders the hydrolysis reaction, which is manifested as a lesser degree of loss in terms of the failure strength.

TGA-DTG Results
The thermal stability of the adhesives in different aging environments was analyzed through TGA-DTG experiments, as shown in Figure 7. From the graph, it can be observed that the degradation process of the adhesive can be roughly divided into the following  Wave Number (cm −1 ) Functional Group In the spectral analysis, the 500-1330 cm −1 segment belongs to the fingerprint region, where the absorption peaks are mainly related to the vibration of a single bond, amino group, and C skeleton, and 1330-4000 cm −1 belongs to the characteristic frequency region, where the absorption peaks are mainly generated by the telescopic vibration of specific functional groups.The absorption peak at 825 cm −1 characterizes the trans stretching of the ether bond to the phenyl group [37].The significant increase in the intensity of the Coatings 2023, 13,1936 13 of 21 1072 cm −1 peak after aging is a strong evidence for the occurrence of the post-curing of the adhesive [34].The absorption peak at 1186 cm −1 represents the stretching of the C-C atoms between two p-phenyls [37], and the characteristic absorption peak of the aromatic band is located at 1339 cm −1 .In a humid and hot environment, the water bound inside the adhesive increases, leading to an increase in the intensity of the absorption peak [38].
The absorption peak at 1700 cm −1 is assigned to the carbon group, the absorption band at 2100 cm −1 is the characteristic absorption peak of the ester group, and the absorption peak at 2850 cm −1 indicates the alkyl group.The intensity of the absorption peaks of the carbon and alkyl groups increases and the intensity of the absorption peaks of the ester group decreases after aging, which indicates that the hydrolysis reaction of the adhesive occurs during the aging process [38].The increase in carbonyl and alkyl groups is due to the hydrolysis of ester groups (R'COOR) under the long-term action of humidity and heat, which indicates that the epoxy resin adhesive may undergo a hydrolysis reaction under the action of humidity and heat aging, and the functional groups are transformed.The hydrolysis reaction formula is as follows: The process in which molecules enter the epoxy resin and react with the corresponding hydrophilic groups is known as hydrolysis, which not only destroys the molecular chain of the polymer in the epoxy resin adhesive, but also destroys the cross-linking of the ester groups, leading to the generation of new carbonyl and alkyl groups.Prolonged moisture and heat attack can lead to changes in the chemical composition of the adhesive, causing irreversible damage to the material.In addition, the lower intensity of the vibrational peaks for carbonyl and alkyl groups with increasing aging time indicates that the post-curing reaction hinders the hydrolysis reaction, which is manifested as a lesser degree of loss in terms of the failure strength.

TGA-DTG Results
The thermal stability of the adhesives in different aging environments was analyzed through TGA-DTG experiments, as shown in Figure 7. From the graph, it can be observed that the degradation process of the adhesive can be roughly divided into the following stages.In the first stage, at around 100-320 • C, the decomposition rate of the adhesive is very slow, and only a small amount of mass reduction (about 7.4%) can be observed.This is mainly due to the moisture in the adhesive and the decomposition of unstable small molecule substances [39].At this stage, some chemical bonds in the adhesive begin to break during the heating process, but the macromolecular substances in the adhesive have not yet broken the chemical bonds.The second stage, at 320~500 • C, in which the decomposition rate is rapid, starting from 320 • C, the sample mass decreases dramatically, which is due to the fact that the adhesive as well as some of the organics start to decompose, especially bromide (bromine is the main constituent element of the adhesive).In the process of increasing the temperature, because the heat gradually penetrates into the adhesive molecules, this leads to the breakage of van der Waals' bonds in the molecules of the brominated organics in the adhesive [14], which results in chain breaks and sidegroup reactions within the polymers.The third stage, at 500~800 • C, requires a very high temperature for complete degradation.The decomposition rate slows down significantly in the third stage, and eventually the quality of the samples with different aging times shows differences, mainly due to two reasons: firstly, because of the evaporation of water from the aged samples, and secondly, because of the degradation of some of the resins already present in the adhesive after the aging process.From the graph, it can be observed that the decomposition temperature of the samples in different aging environments is almost the same; in the salt solution experimental conditions this did not change the thermal stability of the adhesive significantly [40].This can be demonstrated by the final remaining mass.In Figure 7a-c, it can be observed that after aging, the final remaining mass is very close (22%, 23%, and 21%, respectively).
decomposition temperature of the samples in different aging environments is almost the same; in the salt solution experimental conditions this did not change the thermal stability of the adhesive significantly [40].This can be demonstrated by the final remaining mass.In Figure 7a-c, it can be observed that after aging, the final remaining mass is very close (22%, 23%, and 21%, respectively).

Failure Strength Analysis
In order to investigate the effect of moisture and heat aging on the strength of the BFRP-BFRP bonded joints, tensile tests were performed on the bonded joints in three aging environments to obtain the failure strengths of the bonded joints, as shown in Figure 8.In the figure, the downward direction of the arrow represents the decrease of the failure load of the bonded joints.
As can be seen from the figure, the bonded joints in all three environments were significantly damaged.This is due to the fact that the bonded joints are exposed to hot and humid environments, where moisture enters the interior of the adhesive and produces a hydroplasticization effect [18], resulting in a decrease in the failure strength of the bonded joints.Compared with the unaged condition, the failure strength of the bonded joints in the deionized water environment decreased by 18.29%, 27.76%, and 29.17%, respectively, and the bonded joints had a higher failure strength in 3.5% NaCl solution and 5% NaCl solution environment.The failure strength of the bonded joints showed a similar pattern of change, which proves that the NaCl solution inhibits the hygroscopic process of the adhesive, but the strength of this process is not sensitive to the ionic concentration.In addition, no post-curing of the bonded joints was observed in Figure 8, which indicates that the destructive effect of the environment on the bonded joints is greater than the post-curing effect in high temperature and high humidity environments.It was found that regardless of what kind of aging environment the BFRP joints are located in, the degree of failure of the mechanical properties of joints in the first 10 days of aging failure is similar, indicating that the mechanical properties of the BFRP joints are rapidly reduced under the action of humidity-heat coupling, due to the chemical corrosion of the joints caused by salt ions to accelerate the aging of the joints.In the 10-20 day time period of aging failure, the aging rate of the joints in all three environments decreases, and at that stage, the absorption of water is already saturated.The aging rates of the joints in the three environments continued to decrease in the 20-to-30 day time period of aging failure because the joints were in the deposition phase and the post-curing of the adhesive at high temperatures was occurring, so the solution environment was not deterministically related to aging failure during this aging phase.When the aging time limit reached 30 days, the aging degree of the joints in the three environments was similar, but the aging degree of the DW solution environment at 80 • C was greater, indicating that the reduction in the mechanical properties of the joints was positively correlated with the moisture concentration.

Load Displacement Curve
The load displacement curve of the bonded joint before and after aging is shown in Figure 9.The failure load of the bonded joints in the DW environment was 7.129 KN without aging (0 days), and gradually decreased to 5.829 KN, 5.153 KN, and 5.053 KN after 10, 20, and 30 days of aging; the failure load decayed to 5.376 KN after 30 days of aging in 3.5% NaCl solution; the failure load decreased by 2.04 KN after 30 days of aging in 5% NaCl solution, and it decayed by 33.8% after 30 days of aging.After 30 days aging in 3.5% NaCl solution, the failure load decreased to 5.376 KN; in 5% NaCl solution, the failure load decreased by 2.04 KN, which is by 33.8%.The slope of the load-displacement curve indicates the stiffness (ability to resist deformation) of the joint.The difference between the 10-day aged compared to unaged curves is large, and the difference with the aged 20day/30-day curves is small, with the decrease in the magnitude concentrated at the early part of the experiment.This is because the absorption of water is concentrated in the preaging period, the polymer matrix and the adhesive both absorb water and swell to produce internal stresses and partial hydrolysis, which not only reduces the glass transition temperature (Tg), but also reduces the elastic modulus and strength of the adhesive and the substrate.The slopes of the curves show a small increase in the pre-aging period, but then a slight decrease with aging time, indicating that the joints show an increase in stiffness after aging, although the process is extremely slow.The increase in stiffness is due to an initial increase in the secondary cross-linking of the bound water with the polymer

Load Displacement Curve
The load displacement curve of the bonded joint before and after aging is shown in Figure 9.The failure load of the bonded joints in the DW environment was 7.129 KN without aging (0 days), and gradually decreased to 5.829 KN, 5.153 KN, and 5.053 KN after 10, 20, and 30 days of aging; the failure load decayed to 5.376 KN after 30 days of aging in 3.5% NaCl solution; the failure load decreased by 2.04 KN after 30 days of aging in 5% NaCl solution, and it decayed by 33.8% after 30 days of aging.After 30 days aging in 3.5% NaCl solution, the failure load decreased to 5.376 KN; in 5% NaCl solution, the failure load decreased by 2.04 KN, which is by 33.8%.The slope of the load-displacement curve indicates the stiffness (ability to resist deformation) of the joint.The difference between the 10-day aged compared to unaged curves is large, and the difference with the aged 20-day/30-day curves is small, with the decrease in the magnitude concentrated at the early part of the experiment.This is because the absorption of water is concentrated in the pre-aging period, the polymer matrix and the adhesive both absorb water and swell to produce internal stresses and partial hydrolysis, which not only reduces the glass transition temperature (Tg), but also reduces the elastic modulus and strength of the adhesive and the substrate.The slopes of the curves show a small increase in the pre-aging period, but then a slight decrease with aging time, indicating that the joints show an increase in stiffness after aging, although the process is extremely slow.The increase in stiffness is due to an initial increase in the secondary cross-linking of the bound water with the polymer chains, facilitated by the action of moisture and heat, which is subsequently destroyed by the degradation or oxidation of the material.atings 2023, 13, x FOR PEER REVIEW 17 of 22 decreased by 59.24%, 60%, and 52.4%, respectively.This is because after aging, the stiffness of the joint increases and the elastic deformation decreases, resulting in a decrease in the energy absorption effect of the bonded joint [41].The law of change is also with the aging of the failure of the gradual decrease and the magnitude of the decrease in fracture energy is also a large part of the reduction occurs in the aging of the first period, corresponding to the failure of the change in load and the failure of the displacement.The fracture energy was calculated by integrating the load displacement curve, and the changes in the fracture energy before and after the aging of the adhesive joint are shown in Figure 10.In the figure, the blue part represents the unaged bonds, while the right side represents the bonds aged for 10, 20 and 30 days consecutively.The unaged adhesive joints have extremely high energy absorption effects, and the fracture energy of the joints rapidly decreases after aging.The fracture energy of the adhesive joints decreased by 59.24%, 60%, and 52.4%, respectively.This is because after aging, the stiffness of the joint increases and the elastic deformation decreases, resulting in a decrease in the energy absorption effect of the bonded joint [41].The law of change is also with the aging of the failure of the gradual decrease and the magnitude of the decrease in fracture energy is also a large part of the reduction occurs in the aging of the first period, corresponding to the failure of the change in load and the failure of the displacement.

Failure Mode Analysis
The failed sections of the bonded joints after different time aging treatments are shown in Figure 11.SLJ stands for single lap joint.From the figure, it can be seen that bonded joints exhibit mixed failure (tearing failure and cohesive failure) before and after aging, in which the tearing failure is marked by the red curve.When the bonded joint is not aged, the failure mode is dominated by tearing failure.At this point, the interface between the adhesive and the BFRP has a good combining ability.With the extension of the aging time, the water molecules affect the chemical properties of the adhesive and produce damage, so that the bonding ability of the adhesive and the BFRP decreases, and the proportion of tearing of the failed section decreases.In particular, the tear area is very similar in the three environments, especially in the two salt concentrations, which corresponds to the failure strength of the bonded joints.It can be found that after 10 days of aging, the bonded joints in the deionized water environment still have a small portion of tearing, but the salt solution environment bonded joints show a mixed failure dominated by endopolymerization, which indicates that the salt solution has a greater effect on the adhesive than on the BFRP material.This suggests that hot and humid environments as well as salt water environments lead to a reduction in the bond strength of the adhesive.
In addition to the macroscopic failure analysis of fracture surfaces, the failure mechanism of the bonded joints can be further investigated from a fine-scale perspective by scanning electron microscopy (SEM).The SEM test results before and after the aging of the bonded joints are shown in Figure 12.When not aged, the fibers are tightly bound to the resin matrix.When aging proceeds for 10 days, visible cracks can be seen, which is often the case when more moisture is deposited and cracks are produced under load.When aging is in progress for 20 days, the fibers are gradually exposed, with residual torn resin, indicating that the adhesive and resin have suffered some degree of damage here, but the fibers are relatively smooth, a phenomenon that becomes more pronounced after 30 days of aging.Most of the figures exhibit bare fibers, which indicates that the bonding between the adhesive and the BFRP substrate is greater than the bonding between the fibers and the resin.In Figure 12(b3,c2) there are some residual resins; the resin matrix is attached to the surface of the fibers, and the resin matrix is relatively smooth, indicating that cohesive failure is the main form of adhesive joint failure.
resin, indicating that the adhesive and resin have suffered some degree of damage here, but the fibers are relatively smooth, a phenomenon that becomes more pronounced after 30 days of aging.Most of the figures exhibit bare fibers, which indicates that the bonding between the adhesive and the BFRP substrate is greater than the bonding between the fibers and the resin.In Figure 12(b3,c2) there are some residual resins; the resin matrix is attached to the surface of the fibers, and the resin matrix is relatively smooth, indicating that cohesive failure is the main form of adhesive joint failure.

Conclusions
In this paper, the hygroscopicity of the adhesive and the failure mechanism of the bonded joints were investigated by testing BFRP-BFRP bonded joints in different environments.The glass transition temperature of the adhesive was obtained by DSC analysis, and the changes of the functional groups before and after aging as well as the thermal stability of the adhesive were analyzed.Also, the failure strength of the bonded joints be-

Conclusions
In this paper, the hygroscopicity of the adhesive and the failure mechanism of the bonded joints were investigated by testing BFRP-BFRP bonded joints in different environments.The glass transition temperature of the adhesive was obtained by DSC analysis, and the changes of the functional groups before and after aging as well as the thermal stability of the adhesive were analyzed.Also, the failure strength of the bonded joints before and after aging and the failed section were analyzed, informing the durability issues of BFRPs in the automotive manufacturing industry.Based on these results, the following conclusions were drawn: (1) The adhesive has a lower saturated water absorption in the salt solution.The presence of ions inhibits the entry of water into the interior of the adhesive, resulting in a higher failure strength of the bonded joint in the salt solution.This is consistent with the results obtained from the DSC test.
(2) The TGA-DTG results showed that the final residual mass of the adhesive in the three environments was extremely similar, which was 22%, 23%, and 21% of the original mass, respectively.In addition, the TGA-DTG curves of the adhesive after aging for different times were very similar.The decrease in quality is due to the evaporation of water due to the increase in temperature and the degradation of some of the resins in the adhesive.(3) As the aging time increases, The concentration of the salt solution had no significant effect on the adhesive nor on the bonded joints.The Tg values of the adhesive in the two environments of the 3.5% NaCl solution and 5% NaCl solution were similar, and the failure strength curves of the bonded joints in the two environments were close to overlapping.However, the Tg decreased by 46.42% after aging for 30 days in the DW solution environment.The decrease in the adhesive's Tg was related to the humidity-induced plasticizing effect and the decrease in the density of the cross-linked polymer.(4) The bonded joints changed from tearing failure to mixed failure mainly due to internal polymerization before and after aging, and the hydrolysis led to a decrease in the bonding performance of the adhesive, which was corroborated by the FTIR results, which showed that the high temperature decreased the moisture with the increase in the aging time and the post-curing reaction impeded the hydrolysis reaction.The change in the failure mode of the bonded joints is the result of the post-curing effect and the hydrolysis reaction.
It can be concluded from the above that high temperature and salt solution reduce the durability of BFRP bonded joints, but the problem of the harsh temperature environment in which BFRPs are utilized in the automobile manufacturing industry has not yet been fully solved.In this study, only 80 • C was chosen as a temperature condition, and an adhesive was used to study the aging phenomenon of the joints, so there are some limitations to this work.In the future, we will study the effect of different environments on the durability of bonded joints in more depth.

Figure 3 .
Figure 3. (a) The mold of dumbbell specimen is made, (b) dumbbell specimen geometry and dimensions in mm, (c) determinator with an accuracy of 0.1 mg.

Figure 3 .
Figure 3. (a) The mold of dumbbell specimen is made, (b) dumbbell specimen geometry and dimensions in mm, (c) determinator with an accuracy of 0.1 mg.

Figure 5 .
Figure 5. Change in Tg value of adhesive (before and after aging).

Figure 10 .
Figure 10.Fracture energy of a bonded joint.Figure 10.Fracture energy of a bonded joint.

Figure 10 .
Figure 10.Fracture energy of a bonded joint.Figure 10.Fracture energy of a bonded joint.

Table 3 .
Diffusion coefficient and saturation moisture absorption rate.

Table 4 .
Major band assignments in FTIR of adhesives.

Table 4 .
Major band assignments in FTIR of adhesives.